Lathe-DB - integrando um mecanismo de busca por palavras-chave em SGBDs relacionais

Abstract

For many years, several researchers have proposed methods to empower the naive users, without any knowledge about query languages or database details, to perform queries in RDBMS using keywords.ystems that process queries based on keywords in RDBMS, commonly called Relational Keyword-Search - R-KwS , face the task of automatically determine, from some keywords, what information should be retrieved from the database, and how this information can be combined to provide a relevant and quality answer to the user. However a common feature, in the vast majority of current R-KwS systems, is that the implementation of these systems is usually done outside the RDBMS environment. This requires, for its operation, a separate software infrastructure, with complex installation and subsequent hard to perform maintenance. The main objective of this dissertation is integrate a search engine for keyword in a RDBMS, implementing the Lathe method, proposed in the database and information retrieval group of ICOMP, inside PostgreSQL RDBMS environment. Experiments show that the implementation of Lathe-DB, as a search engine inside the RDBMS environment, generate results very close to those of the original work, and that is a very practical, low cost, and adds value, for use the keyword search in RDBMS.